3398
J. Phys. Chem. B 2007, 111, 3398-3403
Aggregative Growth of Silicalite-1 Sandeep Kumar,†,‡ Tracy M. Davis,† Harikrishnan Ramanan,†,‡ R. Lee Penn,*,‡ and Michael Tsapatsis*,† UniVersity of Minnesota, Department of Chemical Engineering and Materials Science, 151 Amundson Hall, 421 Washington AVenue SE, Minneapolis, Minnesota 55455, and UniVersity of Minnesota, Department of Chemistry, B-4, 139 Smith Hall, 207 Pleasant Street SE, Minneapolis, Minnesota 55455 ReceiVed: NoVember 21, 2006; In Final Form: January 26, 2007
Precursor silica nanoparticles can evolve to silicalite-1 crystals under hydrothermal conditions in the presence of tetrapropylammonium (TPA) cations. It has been proposed that in relatively dilute sols of silica, TPA, water, and ethanol, silicalite-1 growth is preceded by precursor nanoparticle evolution and then occurs by oriented aggregation. Here, we present a study of silicalite-1 crystallization in more concentrated mixtures and propose that growth follows a path similar to that taken in the dilute system. Small-angle X-ray scattering (SAXS), cryogenic transmission electron microscopy (cryo-TEM), and high-resolution transmission electron microscopy (HRTEM) were used to measure nanoparticle size and to monitor zeolite nucleation and earlystage crystal development. The precursor silica nanoparticles, present in the clear sols prior to crystal formation, were characterized using two SAXS instruments, and the influence of interparticle interactions is discussed. In addition, SAXS was used to detect the onset of secondary particle formation, and HRTEM was used to characterize their structure and morphology. Cryo-TEM allowed for in situ visual observation of the nanoparticle population. Combined results are consistent with growth by aggregation of silica nanoparticles and of the larger secondary crystallites. Finally, a unique intergrowth structure that was formed during the more advanced growth stages is reported, lending additional support for the proposal of aggregative growth.
Introduction Silicalite-1, an all-silica zeolite with the MFI-type framework,1 has received considerable attention as a model system in the study of hydrothermal crystal growth.2-5 Through a better understanding of zeolite crystallization, it may be possible to tune crystal morphology and size for specialized applications such as in the synthesis of oriented zeolite membranes.6,7 To meet this end, the nucleation and growth behavior of silicalite-1 has been studied at multiple temperatures and compositions.2,8-12 Typically, clear precursor sols are used as the growth medium due to their reduced number of constituents and, therefore, relative simplicity. These clear sols are made from a silica source, most frequently tetraethylorthosilicate (TEOS); the structure directing agent, tetrapropylammonium hydroxide (TPAOH); and water. Figure 1 shows a ternary diagram highlighting a sampling of sol compositions studied by other researchers2,8-12 as well as those considered in this report. Precursor nanoparticles (about 5-nm) spontaneously form in these clear sols immediately following TEOS hydrolysis.11 The structure of the nanoparticles and their role in silicalite-1 crystallization are subjects of extensive research. Researchers differ in their ideas about the relationship of these nanoparticles to the nuclei and building units of silicalite-1. The research group at Leuven13-15 has reported that specific silica oligomers condense to form nanoslabs with the MFI framework, which act as building blocks and directly attach to form silicalite-1. Recently, our group proposed a mechanism2 describing the * Corresponding authors. (Penn) Phone: 612-626-4680, fax: 612-6267541, e-mail:
[email protected]. (Tsapatsis) Phone: 612-626-0920, fax: 612-626-7246, e-mail:
[email protected]. † Department of Chemical Engineering and Materials Science. ‡ Department of Chemistry.
Figure 1. Clear sol compositions considered in previous reports as compared to those from this study.
Figure 2. Schematic illustrating the mechanism proposed by Davis et al.2 for nanoparticle evolution and crystal growth by aggregation (left of the vertical line). The possibility of crystal-crystal aggregation is illustrated on the right.
evolution of precursor nanoparticles to silicalite-1 nuclei and crystal growth by oriented aggregation of these primary particles; a schematic of the proposed pathway is shown in Figure 2. In the schematic, the letter A denotes the precursor nanoparticles with a disordered silica core-TPA shell structure. This species
10.1021/jp0677445 CCC: $37.00 © 2007 American Chemical Society Published on Web 03/09/2007
Aggregative Growth of Silicalite-1 does not contribute to aggregative growth. Rather, the A-species evolve through m intermediates, B1 to Bm, with structure increasingly similar to silicalite-1 nuclei, which are denoted as C1. Species B1 to Bm, although not yet silicalite-1 nuclei, can contribute to crystal growth by attachment, and the increasing subscript of Bm denotes an increasing probability for contributing to growth by aggregation. Growing crystals are denoted as Cn, where n > 2. To date, no technique has been able to elucidate how the nanoparticles structurally evolve as a function of time. However, it is reported that the colloidal stability decreases as a function of time.2 The structurally and functionally distinct populations of nanoparticles of our model explained the presence of relatively large crystals at low yield during the early stages of crystallization as well as the broad size distribution of crystals throughout the growth process.2,16 The possibility of silicalite-1 growth by crystal-crystal aggregation, which, in order to keep the discussion simple, was not considered in the initially proposed mechanism, is now included and illustrated in the schematic (to the right of vertical line in Figure 2). Oriented aggregation occurs when crystals attach at similar crystallographic surfaces to form a new single crystal.17 To date, there are few reports2,14 suggesting the oriented aggregation of precursor nanoparticles during silicalite-1 crystal growth. However, oriented aggregation has been reported to be one of the main mechanisms for nanocrystal growth in various other systems.17-23 Characterization techniques such as small-angle X-ray scattering (SAXS),8,24 dynamic light scattering (DLS),25 nuclear magnetic resonance (NMR) spectroscopy,26,27 and transmission electron microscopy (TEM)3,28,29 have been employed to examine silicalite-1 crystallization in previous studies. SAXS is a uniquely sensitive technique to detect the onset of formation of a secondary population8,24 due to the fact that scattering intensity goes as particle size to the sixth power (intensity ∝ diameter).6 In this way, SAXS is indispensable in studying crystallization from clear sols if the earliest stages of crystal growth are to be captured. However, SAXS provides only an indirect measure of particle size and therefore requires considerable data interpretation. A second technique, for example highresolution transmission electron microscopy (HRTEM), aids this data interpretation while also providing a means for direct observation of crystal evolution at the nanoscale.28-30 The two characterization methods, when used in combination, provide a powerful tool for the elucidation of crystal growth behavior. In the literature, there are few reports of HRTEM investigations of silicalite-1 crystallization possibly due to the severe electron beam damage that occurs during TEM imaging of zeolites.31,32 Furthermore, viewing of the crystals is often obstructed by the more highly abundant amorphous silicate species, especially at low crystal yield. However, Davis et al.2 recently described a two-step dialysis procedure permitting isolation of the silicalite-1 crystals from soluble silica and TPA species. This paper presents a study of crystallization in a relatively concentrated silica-TPA sol and reports a new type of defect in silicalite-1. Experimental observations are made by SAXS, cryo-TEM, and HRTEM. Experimental Methods Synthesis. Silicalite-1 was synthesized from clear precursor sols consisting of TPAOH (1.0 M Aldrich), distilled water, and TEOS (98% Aldrich) at the following molar compositions: 25SiO2:9TPAOH:530H2O:100EtOH (denoted M1) and 25SiO2:3TPAOH:480H2O:100EtOH (denoted M2). Upon mixing of the sol components, M1 was aged at room temperature for 30 days and then heated at 90 °C for various
J. Phys. Chem. B, Vol. 111, No. 13, 2007 3399 times. M2 was characterized immediately after mixing and did not undergo additional heating. Aliquots of M1 were removed at regular time intervals and quenched in a water bath. Small-Angle X-ray Scattering. SAXS was used to characterize the size and concentration of precursor silica nanoparticles and to detect the onset of secondary particle formation. Two instruments were used in the collection of this data. The SAXSess SAXS instrument (Anton-Parr) was employed to characterize fresh M1 and M2 and the aged and hydrothermally treated samples of M1. This instrument is equipped with a Cu KR slit-collimated radiation source. Slit collimation is beneficial in that a larger volume of sample is illuminated at once and therefore data collection time may be significantly reduced without a reduction in measured intensity. However, the scattering profile produced from a line collimated beam will be “smeared” due to the incident beam geometry, an effect that can be accounted for during data analysis using the software program GIFT.33,34 In addition to using the SAXSess instrument, the 5-ID beamline in the DND-CAT Synchrotron Research Center at the Advanced Photon Source, Argonne National Laboratory, with point collimation was used for nanoparticle characterization in M2. Synchrotron data were collected using a CCD detector and then azimuthally integrated to give a onedimensional scattering profile. The scattering vector q was determined from the scattering angle θ through use of the relationship q ) 4πλ-1 sin(θ/2). In all instances, measurements were made at room temperature (25 °C). Cryogenic Transmission Electron Microscopy. Specimens of M1 were prepared without dilution or sol alteration for cryoTEM imaging according to the method described by Talmon.35 Vitrified specimens were then transferred under liquid nitrogen to a Gatan 613.DH cooling holder. Images of the vitrified sol (T < -170 °C) were taken with a JEOL 1210 TEM operating at 120 kV and collected using a Gatan charge-couple device (CCD) camera. Transmission Electron Microscopy. The two-step dialysis procedure described by Davis et al.2 was performed to isolate silicalite-1 crystals from precursor nanoparticles, dissolved silica, and unreacted TPA cations prior to sample preparation for HRTEM. The purpose of dialysis was to minimize silica and TPA precipitation that typically occurs upon specimen drying. TEM specimens were prepared by diluting a small amount of dialyzed sol in ethanol and placing a few droplets onto a holey carbon-coated copper grid (Ted Pella, Inc.). Samples were then allowed to air-dry. TEM characterization was performed using a FEI Tecnai G2 F30 TEM operating at 300 kV. All TEM micrographs were collected using a CCD camera. Results and Discussion Characterization of Clear Sols. Particles ranging in size from about 1 to 50 nm and suspended in a medium (e.g., silica nanoparticles in water) are easily detectable by small-angle scattering techniques given that the contrast between the particle and matrix is sufficient. In the case of X-ray radiation, the contrast is defined as the electron density difference between the particle and matrix. The measured scattering intensity, collected as a function of the scattering vector, q, is a function of this contrast as well as a function of the particle volume, V, and number concentration, n, according to the following relationship:36
I(q) ) nV2 contrast2 P(q) S(q) The form factor, P(q), results from intraparticle scattering and gives information about particle size and shape. The
3400 J. Phys. Chem. B, Vol. 111, No. 13, 2007
Kumar et al.
Figure 4. TEM images of silicalite-1 crystals from sol heat-treated for 52 h at 90 °C. Rounded protrusions of approximately 5-nm size are shown by arrowheads in (a) and (c).
Figure 3. SAXS intensity profiles of M1 and M2. (a) Characterization of precursor nanoparticles in M2 showing comparison of desmeared (I) and line-smeared (II) SAXSess data to point-smeared Synchrotron (III) data. The inset gives the PDDF determined from the desmeared SAXSess data. (b) Measurements from M1 with increasing time steps: immediately after TEOS hydrolysis (t1), after 30 days at room temperature (t2), and 30 days at room temperature followed by 2 h (t3), 3 h (t4), and 6 h at 90 °C (t5). The onset of secondary particle formation in M1 is shown to occur during the first 2 h at 90 °C. M1 prior to aging was also characterized by cryo-TEM (c).
structure factor, S(q), is a result of interparticle scattering and therefore describes particle-particle interactions and the spatial arrangement of the particles in the solvent. In the case of dilute sols, S(q) is reduced to unity and the problem of data analysis is greatly simplified. Scattering profiles from M1 and M2 (shown in Figure 3) were collected in order to characterize the precursor nanoparticles and to monitor the evolution of the sols as a function of time and temperature. SAXS data collected from fresh M2 using the SAXSess and the Synchrotron source at Argonne are shown in Figure 3a. Both the line-smeared and the desmeared SAXSess data are given. Comparison of these two profiles shows the influence of incident beam geometry and the impact of data smearing on profile appearance. The peak at q ) 0.58 nm-1, which is easily visible in the desmeared SAXSess data as well as in the Synchrotron data, is a direct result of interparticle interactions (i.e., the structure factor). The position of this peak maximum gives information about the average particle separation and is often interpreted as the effective particle size (d) through the relationship: d ) 2π/qmax. Accordingly, the size of the precursor nanoparticles in M2 can be approximated to be 10.8 nm (i.e., 2π/0.58 nm-1). However, a more accurate measure of the average particle size is obtained by taking the indirect Fourier transform of the desmeared SAXS data, yielding the pair distance distribution function (PDDF). This manipulation of the data has been performed through the use of the software program GIFT.33,34 The resultant PDDF (shown in the inset of Figure 3a) gives a maximum nanoparticle dimension of 8.5 nm. The discrepancy between the particle size as determined from the PDDF and that calculated from the SAXS peak position is a direct result of the interparticle interactions. The evolution of the clear sols of composition M1 was monitored using the SAXSess instrument and is shown in Figure 3b. Measurements were taken at the following time steps: immediately after TEOS hydrolysis (t1), after 30 days at room
temperature (t2), and 30 days at room temperature followed by 2 h (t3), 3 h (t4), and 6 h at 90 °C (t5). The diameter of the precursor nanoparticles at t1 as determined from the PDDF (not shown) is ∼4.0 nm and agrees well with cryo-TEM imaging (Figure 3c). The significant size difference between particles in M1 and M2 may be due to the difference in SiO2:TPA; an explanation of this particle-size dependence was previously given by Yang et al.12 The precursor nanoparticles in M1 increase in size to ∼4.7 nm and exist as the only detectable species during the 30 days at room temperature. However, after only 2 h at elevated temperature, a secondary population of larger-sized particles is detected, causing a sharp increase in the low-q scattering intensity. As suggested by modeling to be reported elsewhere,37 concentrations undetectable by standard physical measurements (e.g., weight, X-ray diffraction), corresponding to crystal yields at a fraction of 1%, are easily detected by SAXS. Immediately following detection of the larger particles, aliquots were removed for dialysis and HRTEM characterization. The scattering intensity continues to increase beyond the initial detection of the secondary particles, signifying an increase in their size and number. Aggregative Crystal Growth. HRTEM results show that the secondary particles are crystals of silicalite-1, which is consistent with previous results in the more dilute system.2 The silicalite-1 crystallites produced by heating M1 at 90 °C for 52 h are typically faceted and have the hexagonal prismatic shape that is characteristic of this zeolite phase.38-40 Figure 4 shows representative TEM images of crystallites that were isolated by the two-step dialysis procedure.2 The vast majority of the crystallites display the well-faceted appearance of Figure 4, parts a and b, which show silicalite-1 crystallites oriented along [010]. Upon close inspection, however, small deviations from the hexagonal prismatic shape can be discerned. For example, in Figure 4a there are rounded protrusions that are approximately 5 nm in size (illustrated by arrowheads). Similar features are evident in most images, including those of crystallites in other orientations (e.g., Figure 4c). In Figure 4c, crystal dimensions along [100] and [010] are comparable, which is expected for silicalite-1.38-40 Interestingly, such features were also evident in crystals produced from a much more dilute sol (H2O:SiO2 about 22 times greater than in present work).2 A preliminary explanation for the protrusions is the adhesion of nanoparticles (ca. 5 nm) to the crystals. While in situ AFM imaging provided a second means for detecting nanoparticle adsorption on etched mica in the more dilute system, data interpretation in the presence of silicalite-1 crystals is less transparent. Specifically, distinction between silicalite-1 surface features and nanoparticles adsorbed on the crystal surface is difficult by in situ AFM and
Aggregative Growth of Silicalite-1
J. Phys. Chem. B, Vol. 111, No. 13, 2007 3401
Figure 5. TEM image of silicalite-1 crystal from sol heat-treated for 52 h at 90 °C.
Figure 6. TEM images ((a) and (b)) of the first evidence of crystallinity in silicalite-1 present in sol after heat-treatment for ∼2 h at 90 °C. Silicalite-1 in (a) is oriented either along [100] or [010]; FFTs are shown in insets. TEM images on the left are reproduced on the right, and broken white lines serve to highlight the perimeters of the crystallites.
does not provide additional information in regard to the observed 5-nm protrusions shown in Figure 4a. Some crystallites exhibit faceted morphologies that not only have the small rounded features but also are quite asymmetric as shown in Figure 5. This figure shows a morphology that suggests growth occurred through the oriented aggregation of at least two silicalite-1 crystals of sizes 60 × 40 and 40 × 30 nm.17,18 As the SAXS results suggest, the emergence of secondary particles (with a larger size) occurs much earlier than 52 h. In fact, HRTEM and SAXS data clearly demonstrate that silicalite-1 crystallites have formed within 2 h at 90 °C. Through time-resolved HRTEM characterization, crystal development in M1, progressing from the observation of extremely small crystallites (ca. 10 nm) to relatively large crystals (ca. 50 nm, as shown in Figures 4 and 5) was captured. Figures 6-8 show TEM images of crystallites isolated from suspensions heat-treated at 90 °C for 2, 3, and 6 h, respectively. No evidence of crystal lattice fringes was seen by HRTEM prior to heat-treatment at 90 °C. In addition, electron diffraction patterns (not shown) performed on cryo-TEM specimens (e.g., Figure 3c) are characteristic of an amorphous material prior to heat-treatment. Crystallinity was first observed by HRTEM in samples prepared after 2 h at 90 °C. Representative TEM images showing ∼10-nm crystallites are shown in Figure 6; FFTs are shown in the insets. Figure 6a shows a TEM image of a silicalite-1 crystallite oriented either along [100] or [010]. Figure
Figure 7. TEM images of silicalite-1 crystals present in sol after heattreatment for 3 h at 90 °C; FFTs are shown in insets. Crystal in (a) is oriented along [111]-axis while it is oriented either along [100] or [010] in (b).
Figure 8. TEM images showing silicalite-1 crystals formed after heattreatment for 6 h at 90 °C.
6b shows a TEM image of a crystallite oriented away from any major zone axis with a lattice fringe spacing consistent with the MFI-type framework. The images in Figure 6, parts a and b, are reproduced on the right, and a dashed white line serves to highlight the perimeter of the crystallites. The crystallites appear to be composed of rounded silicalite-1 subdomains in the approximately 5-10-nm size range. The rounded features evident in the images shown in Figure 4 are also in this size range. Figure 7 shows TEM images of silicalite-1 after heattreatment at 90 °C for 3 h; crystals have increased by as much
3402 J. Phys. Chem. B, Vol. 111, No. 13, 2007
Kumar et al.
Figure 10. TEM images showing intergrown silicalite-1 crystals present in sol after heat-treatment for 52 h at 90 °C. Pairs of white arrows show the crystallographic orientation of the two regions.
Figure 9. TEM images of silicalite-1 crystals present in sol after heattreatment for 6 h at 90 °C; FFTs are shown in insets.
as 15 nm in size. The silicalite-1 crystal shown in Figure 7a is oriented along the [111]-axis, and the crystal imaged in Figure 7b is oriented close to the [100]- or the [010]- axis; FFTs are shown in the insets. Figures 8 and 9 show representative TEM images of silicalite-1 crystals present after heat-treatment at 90 °C for 6 h. The crystals exhibit aggregate-like morphologies. Magnified images of two crystals in Figure 8, indicated by white arrows numbered 1 and 2, are shown in the insets. Crystals were observed to be ∼10-20-nm wide and ∼20-50-nm long with lattice fringes consistent with the MFI-type framework. In crystal 1, there is a slight misorientation between the lower portion and the remainder of the crystallite. This misorientation is highlighted by the white lines, which are drawn parallel to local lattice fringes and have an angular mismatch of ∼19°. Crystal 2 is an aggregate-like cluster composed of ∼5-nm-sized nanocrystalline domains (indicated by the arrowheads). Similar features were evident in most of the imaged silicalite-1 crystals. Such aggregate-like crystals with domain sizes on the order of the precursor nanoparticles (ca. 5 nm) are suggestive of crystal growth by oriented aggregation.2,4,17,18,41,42 The observed crystal morphology is very different from the compact morphology reported for the aluminosilicate zeolites, for which a gel-tozeolite transformation has been proposed.28,29 The presence of smaller crystallites, frequently observed with the larger crystals, provides additional evidence of the oriented aggregation growth mechanism. An example image illustrating the broad crystal size distribution is given in Figure 9a; the top crystal is ∼10 nm, whereas neighboring crystals are ∼20 nm in size. A wide particle size distribution is consistent with the model proposed
previously2,16 for silicalite-1 crystal growth in dilute silica sols. Although aggregation of amorphous nanoparticles followed by crystallization of the aggregate upon reaching some critical aggregate size is another possibility to explain the observed morphology, a tighter particle size distribution would be expected and the observed lattice misorientations could not be explained. It is also possible that the dialysis procedure alters the crystal morphology in some way. For example, amorphous regions of the particle may dissolve upon dialysis. However, the formation of single-crystal aggregates during dialysis is improbable, and therefore oriented aggregation remains as the most likely explanation of the experimental observations reported here. Detailed examination of these silicalite-1 images (Figures 6-8) yields three key observations. First, crystallite size increases with increasing crystallization time at 90 °C. Second, the crystallites appear to be composed of an ever-increasing number of primary particles. Third, the rounded protrusions are retained and are in the approximately 5-10-nm size range. On the basis of these observations, we hypothesize that the rate of growth by aggregation during the early stages of crystallization is substantially faster than the rate of recrystallization, a process that would smooth out such features. During the latter stages of growth (i.e., 52 h), the crystals become more compact and faceted, suggesting that growth at these longer times is occurring by monomer addition. This competing growth process (monomer addition) becomes important during the latter stages due to a combination of increased crystal surface area, increased sol pH, and reduced nanoparticle number concentration and was also observed in the more dilute system.2 Consideration of the timeresolved SAXS profiles (Figure 3b) provides evidence that after 6 h at 90 °C the number of nanoparticles present has been reduced considerably. More specifically, the scattering intensity for q > 0.8 nm-1 is significantly less after the 6-hour heattreatment than the scattering intensity at earlier times. An additional feature consistent with the proposed growth mechanism persists in some crystals up to at least 52 h of hydrothermal growth. Specifically, silicalite-1 crystals imaged after 52 h at 90 °C are sometimes found to contain unique defects that may be explained by an aggregative growth mechanism. Silicalite-1 crystals possessing this signature defect are shown in Figure 10, parts a-c. The images show silicalite-1 crystals composed of distinct regions that are misoriented with respect to one another. Moreover, the individual crystalline regions have a common axis of rotation , and for the crystals shown in Figure 10, the regions are related to each other
Aggregative Growth of Silicalite-1 by an angle of 37.3° ((2.5). The measured angle of rotation was determined from calibrated images using Gatan Digital Micrograph. To date, this type of misorientation in silicalite-1 has not been reported. Further inspection shows that the interface between the misoriented regions is not a sharp boundary (Figure 10, parts a and c). The measured angles of misorientation, ∼37.3° for the three crystals shown in Figure 10, may be interpreted by considering the silicalite-1 crystal structure (MFI-type framework). In this zeolite, the angle between and in the MFI framework is ∼34°. In the image of the silicalite-1 crystal shown in Figure 10a, [101]A is parallel to [001]B, and the misorientation between [101]A and [101]B is 37°. This data could indicate that it is the arrangement of the straight pores that controls the nature of the defect. The angular consistency coupled with the morphology of these intergrown crystals suggests that such particles are formed by oriented aggregation. If, instead, the intergrowth was the result of two crystallization events in a single amorphous particle, the angular relationship would likely vary from one intergrowth to the next. There are reports that growth by oriented aggregation can lead to the formation of defects such as twin boundaries and other interfaces.2,17,41 The only requirement for oriented aggregation is that structural accord be achieved at the interfaces. Thus, we hypothesize that when silicalite-1 crystallites achieve an orientation in which the straight pores in one subunit are arranged parallel to the [001] in a second subunit, oriented aggregation can occur. Preliminary results are in agreement with this argument; however, more results and extensive image simulations are necessary to fully interpret the dependence of the intergrown particles’ angular relationship on the arrangement of straight pores in silicalite-1. The low frequency of intergrown particles likely indicates that this type of structural accord does not result in as low an energy configuration as oriented aggregation in which the crystallites are oriented with respect to all three crystallographic axes (or close to it). It should also be noted that intergrown silicalite-1 crystals were not observed during the early stages of crystallization. Rather, it seems that the intergrowth is a unique consequence of crystal-crystal aggregation (Cn-Cn) during the latter stages. Conclusions SAXS and time-resolved HRTEM were employed to track silicalite-1 crystallization in a concentrated silica sol. SAXS and cryo-TEM provided a means for characterizing the precursor nanoparticles, and SAXS enabled early detection of larger-sized secondary particles, samples of which could then be prepared for HRTEM imaging. The development and growth of aggregate-like crystals (Figures 6-9) that later transform to display faceted (Figure 4) to rounded-faceted morphologies (Figures 5 and 10) was followed by HRTEM. Aggregate-like crystals with domains on the order of the precursor nanoparticles during the initial stage of crystallization are consistent with the oriented aggregation mechanism. The reported intergrowth structures observed in the silicalite-1 crystals provide additional evidence for growth by oriented aggregation. The results from this hydrothermal growth study of silicalite-1, when taken collectively, strongly suggest that crystallization in a concentrated precursor sol follows the same pathway (recall schematic in Figure 2) that was proposed to describe crystal growth in a more dilute system.2 Acknowledgment. Financial support for this work was provided by the MRSEC Program of NSF under Award Number
J. Phys. Chem. B, Vol. 111, No. 13, 2007 3403 DMR-0212302 and in part by NSF Awards NIRT CTS0103010, CTS-0522518, and NSF-MRI EAR-0320641. Characterization was carried out at the Characterization Facility, University of Minnesota, which receives support from NSF through the National Nanotechnology Infrastructure Network. References and Notes (1) Database of the Structure Commission of the International Zeolite Association (http://www.iza-structure.org/databases/), 2006. (2) Davis, T. M.; Drews, T. O.; Ramanan, H.; He, C.; Dong, J.; Schnablegger, H.; Katsoulakis, M. A.; Kokkoli, E.; McCormick, A. V.; Penn, R. L.; Tsapatsis, M. Nat. Mater. 2006, 5, 400. (3) Mintova, S.; Olson, N. H.; Senker, J.; Bein, T. Angew. Chem., Int. Ed. 2002, 41, 2558. (4) Nikolakis, V.; Kokkoli, E.; Tirrell, M.; Tsapatsis, M.; Vlachos, D. G. Chem. Mater. 2000, 12, 845. (5) Schoeman, B. J. Microporous Mesoporous Mater. 1998, 22, 9. (6) Choi, J.; Ghosh, S.; Lai, Z.; Tsapatsis, M. Angew. Chem., Int. Ed. 2006, 45, 1154. (7) Lai, Z.; Tsapatsis, M.; Nicolich, J. P. AdV. Funct. Mater. 2004, 14, 716. (8) De Moor, P. E. A.; Beelen, T. P. M.; Santen, R. A. V.; Beck, L. W.; Davis, M. E. J. Phys. Chem. B 2000, 104, 7600. (9) Fedeyko, J. M.; Rimer, J. D.; Lobo, R. F.; Vlachos, D. G. J. Phys. Chem. B 2004, 108, 12271. (10) Houssin, C. J. Y.; Kirschhock, C. E. A.; Magusin, C. M. M.; Mojet, B. L.; Grobet, P. J.; Jacobs, P. A.; Martens, J. A.; Santen, R. A. V. Phys. Chem. Chem. Phys. 2003, 5, 3518. (11) Schoeman, B. J.; Regev, O. Zeolites 1996, 17, 447. (12) Yang, S.; Navrotsky, A. Chem. Mater. 2004, 16, 210. (13) Kirschhock, C. E. A.; Kremer, S. P. B.; Vermant, J.; Tendeloo, G. V.; Jacobs, P. A.; Martens, J. A. Chem. Eur. J. 2005, 11, 4306. (14) Kirschhock, C. E. A.; Ravishankar, R.; Jacobs, P. A.; Martens, J. A. J. Phys. Chem. B 1999, 103, 11021. (15) Kirschhock, C. E. A.; Buschmann, V.; Kremer, S.; Ravishankar, R.; Houssin, C. J. Y.; Mojet, B. L.; van Santen, R. A.; Grobert, P. J.; Jacobs, P. A.; Martens, J. A. Angew. Chem., Int. Ed. 2001, 40, 2637. (16) Drews, T. O.; Tsapatsis, M. Microporous Mesoporous Mater., in press. (17) Penn, R. L.; Banfield, J. F. Am. Mineral. 1998, 83, 1077. (18) Penn, R. L.; Banfield, J. F. Geochim. Cosmochim. Acta 1999, 63, 1549. (19) Banfield, J. F.; Welch, S. A.; Zhang, H.; Ebert, T. T.; Penn, R. L. Science 2000, 289, 751. (20) Liu, Z.; Sakamoto, Y.; Ohsuna, T.; Hiraga, K.; Terasaki, O.; Ko, C. H.; Shin, H. J.; Ryoo, R. Angew. Chem., Int. Ed. 2000, 39, 3107. (21) Lee, E. J. H.; Ribeiro, C.; Longo, E.; Leite, E. R. J. Phys. Chem. B 2005, 109, 20842. (22) Ribeiro, C.; Lee, E. J. H.; Longo, E.; Leite, E. R. Chem. Phys. Chem. 2005, 6, 690. (23) Gehrke, N.; Colfen, H.; Pinna, N.; Antonietti, M.; Nassif, N. Cryst. Growth Des. 2005, 5, 1317. (24) De Moor, P. E. A.; Beelen, T. P. M.; Santen, R. A. V.; Tsuji, K.; Davis, M. E. Chem. Mater. 1999, 11, 36. (25) Schoeman, B. J. Zeolites 1997, 18, 97. (26) Kirschhock, C. E. A.; Ravishankar, R.; Verspeurt, F.; Grobert, P. J.; Jacobs, P. A.; Martens, J. A. J. Phys. Chem. B 1999, 103, 4965. (27) Burkett, S. L.; Davis, M. E. J. Phys. Chem. 1994, 98, 4647. (28) Mintova, S.; Olson, D. H.; Bein, T. Angew. Chem., Int. Ed. 1999, 38, 3201. (29) Mintova, S.; Olson, D. H.; Valtchev, V.; Bein, T. Science 1999, 283, 958. (30) Tsapatsis, M.; Lovallo, M.; Davis, M. E. Microporous Mater. 1996, 5, 381. (31) Bursill, L. A.; Thomas, J. M.; Rao, K. J. Nature 1981, 289, 157. (32) Pan, M. Micron 1996, 27, 219. (33) Bergmann, A.; Fritz, G.; Glatter, O. J. Appl. Cryst. 2000, 33, 1212. (34) Glatter, O. J. Appl. Cryst. 1979, 12, 166. (35) Talmon, Y. Colloids Surf. 1986, 19, 237. (36) Glatter, O.; Kratky, O. Small Angle X-ray Scattering; Academic Press Inc. Ltd.: London, 1982. (37) Davis, T. M.; Drews, T. O.; Tsapatsis, M. To be submitted. (38) Burchart, E. D.; Jansen, J. C.; Graaf, B. V.; Bekkum, H. V. Zeolites 1993, 13, 216. (39) Diaz, I.; Kokkoli, E.; Terasaki, O.; Tsapatsis, M. Chem. Mater. 2004, 16, 5226. (40) Bonilla, G.; Diaz, I.; Tsapatsis, M.; Jeong, H.-K.; Lee, Y.; Vlachos, D. G. Chem. Mater. 2004, 16, 5697. (41) Penn, R. L.; Banfield, J. F. Science 1998, 281, 969. (42) Penn, R. L. J. Phys. Chem. B 2004, 108, 12707.
